maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Transforms/ObjCARC/invoke.ll

225 lines
6.7 KiB
LLVM
Raw Normal View History

Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; RUN: opt -S -objc-arc < %s | FileCheck %s
declare i8* @objc_retain(i8*)
declare void @objc_release(i8*)
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
declare i8* @objc_retainAutoreleasedReturnValue(i8*)
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
declare i8* @objc_msgSend(i8*, i8*, ...)
declare void @use_pointer(i8*)
declare void @callee()
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
declare i8* @returner()
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; ARCOpt shouldn't try to move the releases to the block containing the invoke.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test0(
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: invoke.cont:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK: call void @objc_release(i8* %zipFile) [[NUW:#[0-9]+]], !clang.imprecise_release !0
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: ret void
; CHECK: lpad:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK: call void @objc_release(i8* %zipFile) [[NUW]], !clang.imprecise_release !0
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: ret void
[objc-arc] Test cleanups. Mainly adding paranoid checks for the closing brace of a function to help with FileCheck error readability. Also some other minor changes. No actual CHECK changes. llvm-svn: 180668
2013-04-27 13:25:54 +08:00
; CHECK-NEXT: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test0(i8* %zipFile) personality i32 (...)* @__gxx_personality_v0 {
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
entry:
call i8* @objc_retain(i8* %zipFile) nounwind
call void @use_pointer(i8* %zipFile)
invoke void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*)*)(i8* %zipFile)
to label %invoke.cont unwind label %lpad
invoke.cont: ; preds = %entry
call void @objc_release(i8* %zipFile) nounwind, !clang.imprecise_release !0
ret void
lpad: ; preds = %entry
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%exn = landingpad {i8*, i32}
Update more tests to the new EH scheme. llvm-svn: 138894
2011-09-01 05:04:11 +08:00
cleanup
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
call void @objc_release(i8* %zipFile) nounwind, !clang.imprecise_release !0
ret void
}
; ARCOpt should move the release before the callee calls.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test1(
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: invoke.cont:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK: call void @objc_release(i8* %zipFile) [[NUW]], !clang.imprecise_release !0
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: call void @callee()
; CHECK: br label %done
; CHECK: lpad:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK: call void @objc_release(i8* %zipFile) [[NUW]], !clang.imprecise_release !0
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
; CHECK: call void @callee()
; CHECK: br label %done
; CHECK: done:
; CHECK-NEXT: ret void
[objc-arc] Test cleanups. Mainly adding paranoid checks for the closing brace of a function to help with FileCheck error readability. Also some other minor changes. No actual CHECK changes. llvm-svn: 180668
2013-04-27 13:25:54 +08:00
; CHECK-NEXT: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test1(i8* %zipFile) personality i32 (...)* @__gxx_personality_v0 {
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
entry:
call i8* @objc_retain(i8* %zipFile) nounwind
call void @use_pointer(i8* %zipFile)
invoke void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*)*)(i8* %zipFile)
to label %invoke.cont unwind label %lpad
invoke.cont: ; preds = %entry
call void @callee()
br label %done
lpad: ; preds = %entry
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%exn = landingpad {i8*, i32}
Update more tests to the new EH scheme. llvm-svn: 138894
2011-09-01 05:04:11 +08:00
cleanup
Fix ARCOpt to insert releases on both successors of an invoke rather than trying to insert them immediately after the invoke. llvm-svn: 133188
2011-06-17 04:57:14 +08:00
call void @callee()
br label %done
done:
call void @objc_release(i8* %zipFile) nounwind, !clang.imprecise_release !0
ret void
}
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
; The optimizer should ignore invoke unwind paths consistently.
; PR12265
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
; CHECK: define void @test2() personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
; CHECK: invoke.cont:
; CHECK-NEXT: call i8* @objc_retain
Fix broken check lines. I really need to find a way to automate this, but I can't come up with a regex that has no false positives while handling tricky cases like custom check prefixes. llvm-svn: 162097
2012-08-17 20:28:26 +08:00
; CHECK-NOT: @objc_r
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
; CHECK: finally.cont:
; CHECK-NEXT: call void @objc_release
Fix broken check lines. I really need to find a way to automate this, but I can't come up with a regex that has no false positives while handling tricky cases like custom check prefixes. llvm-svn: 162097
2012-08-17 20:28:26 +08:00
; CHECK-NOT: @objc
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
; CHECK: finally.rethrow:
Fix broken check lines. I really need to find a way to automate this, but I can't come up with a regex that has no false positives while handling tricky cases like custom check prefixes. llvm-svn: 162097
2012-08-17 20:28:26 +08:00
; CHECK-NOT: @objc
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
; CHECK: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test2() personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
entry:
%call = invoke i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* ()*)()
to label %invoke.cont unwind label %finally.rethrow, !clang.arc.no_objc_arc_exceptions !0
invoke.cont: ; preds = %entry
%tmp1 = tail call i8* @objc_retainAutoreleasedReturnValue(i8* %call) nounwind
call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void ()*)(), !clang.arc.no_objc_arc_exceptions !0
invoke void @use_pointer(i8* %call)
to label %finally.cont unwind label %finally.rethrow, !clang.arc.no_objc_arc_exceptions !0
finally.cont: ; preds = %invoke.cont
tail call void @objc_release(i8* %call) nounwind, !clang.imprecise_release !0
ret void
finally.rethrow: ; preds = %invoke.cont, %entry
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%tmp2 = landingpad { i8*, i32 }
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
catch i8* null
unreachable
}
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; Don't try to place code on invoke critical edges.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test3(
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK: if.end:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK-NEXT: call void @objc_release(i8* %p) [[NUW]]
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK-NEXT: ret void
[objc-arc] Test cleanups. Mainly adding paranoid checks for the closing brace of a function to help with FileCheck error readability. Also some other minor changes. No actual CHECK changes. llvm-svn: 180668
2013-04-27 13:25:54 +08:00
; CHECK-NEXT: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test3(i8* %p, i1 %b) personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
entry:
%0 = call i8* @objc_retain(i8* %p)
call void @callee()
br i1 %b, label %if.else, label %if.then
if.then:
invoke void @use_pointer(i8* %p)
to label %if.end unwind label %lpad, !clang.arc.no_objc_arc_exceptions !0
if.else:
invoke void @use_pointer(i8* %p)
to label %if.end unwind label %lpad, !clang.arc.no_objc_arc_exceptions !0
lpad:
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%r = landingpad { i8*, i32 }
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
cleanup
ret void
if.end:
call void @objc_release(i8* %p)
ret void
}
; Like test3, but with ARC-relevant exception handling.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test4(
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK: lpad:
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
; CHECK-NEXT: %r = landingpad { i8*, i32 }
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK-NEXT: cleanup
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK-NEXT: call void @objc_release(i8* %p) [[NUW]]
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK-NEXT: ret void
; CHECK: if.end:
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK-NEXT: call void @objc_release(i8* %p) [[NUW]]
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
; CHECK-NEXT: ret void
[objc-arc] Test cleanups. Mainly adding paranoid checks for the closing brace of a function to help with FileCheck error readability. Also some other minor changes. No actual CHECK changes. llvm-svn: 180668
2013-04-27 13:25:54 +08:00
; CHECK-NEXT: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test4(i8* %p, i1 %b) personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
entry:
%0 = call i8* @objc_retain(i8* %p)
call void @callee()
br i1 %b, label %if.else, label %if.then
if.then:
invoke void @use_pointer(i8* %p)
to label %if.end unwind label %lpad
if.else:
invoke void @use_pointer(i8* %p)
to label %if.end unwind label %lpad
lpad:
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%r = landingpad { i8*, i32 }
It's not possible to insert code immediately after an invoke in the same basic block, and it's not safe to insert code in the successor blocks if the edges are critical edges. Splitting those edges is possible, but undesirable, especially on the unwind side. Instead, make the bottom-up code motion to consider invokes to be part of their successor blocks, rather than part of their parent blocks, so that it doesn't push code past them and onto the edges. This fixes PR12307. llvm-svn: 153343
2012-03-24 01:47:54 +08:00
cleanup
call void @objc_release(i8* %p)
ret void
if.end:
call void @objc_release(i8* %p)
ret void
}
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
; Don't turn the retainAutoreleaseReturnValue into retain, because it's
; for an invoke which we can assume codegen will put immediately prior.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test5(
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
; CHECK: call i8* @objc_retainAutoreleasedReturnValue(i8* %z)
; CHECK: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test5() personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
entry:
%z = invoke i8* @returner()
to label %if.end unwind label %lpad, !clang.arc.no_objc_arc_exceptions !0
lpad:
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%r13 = landingpad { i8*, i32 }
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
cleanup
ret void
if.end:
call i8* @objc_retainAutoreleasedReturnValue(i8* %z)
ret void
}
; Like test5, but there's intervening code.
Catch more CHECK that can be converted to CHECK-LABEL in Transforms for easier debugging. No functionality change. This conversion was done with the following bash script: find test/Transforms -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_]*\):\( *\)define\([^@]*\)@$FUNC\([( ]*\)\$/;\1\2-LABEL:\3define\4@$FUNC(/g" $TEMP done mv $TEMP $NAME fi done llvm-svn: 186269
2013-07-14 09:50:49 +08:00
; CHECK-LABEL: define void @test6(
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
; CHECK: call i8* @objc_retain(i8* %z)
; CHECK: }
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
define void @test6() personality i8* bitcast (i32 (...)* @__objc_personality_v0 to i8*) {
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
entry:
%z = invoke i8* @returner()
to label %if.end unwind label %lpad, !clang.arc.no_objc_arc_exceptions !0
lpad:
Move the personality function from LandingPadInst to Function The personality routine currently lives in the LandingPadInst. This isn't desirable because: - All LandingPadInsts in the same function must have the same personality routine. This means that each LandingPadInst beyond the first has an operand which produces no additional information. - There is ongoing work to introduce EH IR constructs other than LandingPadInst. Moving the personality routine off of any one particular Instruction and onto the parent function seems a lot better than have N different places a personality function can sneak onto an exceptional function. Differential Revision: http://reviews.llvm.org/D10429 llvm-svn: 239940
2015-06-18 04:52:32 +08:00
%r13 = landingpad { i8*, i32 }
Don't convert objc_retainAutoreleasedReturnValue to objc_retain if it is retaining the return value of an invoke that it immediately follows. llvm-svn: 153344
2012-03-24 02:09:00 +08:00
cleanup
ret void
if.end:
call void @callee()
call i8* @objc_retainAutoreleasedReturnValue(i8* %z)
ret void
}
Update more tests to the new EH scheme. llvm-svn: 138894
2011-09-01 05:04:11 +08:00
declare i32 @__gxx_personality_v0(...)
When an invoke is marked with metadata indicating its unwind edge should be ignored by ARC optimization, don't insert new ARC runtime calls in the unwind destination. llvm-svn: 152748
2012-03-15 07:05:06 +08:00
declare i32 @__objc_personality_v0(...)
Update more tests to the new EH scheme. llvm-svn: 138894
2011-09-01 05:04:11 +08:00
Use references to attribute groups on the call/invoke instructions. Listing all of the attributes for the callee of a call/invoke instruction is way too much and makes the IR unreadable. Use references to attributes instead. llvm-svn: 175877
2013-02-22 17:09:42 +08:00
; CHECK: attributes [[NUW]] = { nounwind }
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!0 = !{}